{"id":"cx-decision-tree-actions-four-cases","audience":"customer-support","category":"support-work","technique":"constraints","track":"support","level":"intermediate","title":"Apply a decision tree in order to four ticket cases","free":false,"task":"Apply the decision tree to the four cases. Exactly four lines formatted \"Case 1: Action\", where Action is exactly one of Resolve now, Escalate to Tier 2, Ask customer for info, or Close as duplicate. Apply the checks in the order given and stop at the first match. Plain text, nothing else.","materialLabel":"Your material (attached to your prompt automatically)","material":"TRIAGE DECISION TREE (check in this order; stop at the first match)\n1. The customer already has an open ticket for the same issue -> Close as duplicate.\n2. Key information is missing (no order number, no screenshot or no error text) -> Ask customer for info.\n3. A help-center article answers the question -> Resolve now (send the article).\n4. An error code appears and the customer already tried restarting -> Escalate to Tier 2.\n\nCASES\nCase 1: A customer sends the same login problem again. Ticket T-55 for it is still open.\nCase 2: A customer writes 'it's broken' with no details.\nCase 3: A customer asks how to reset their password, with their account email. The reset article exists.\nCase 4: A customer reports error E-409 after restarting twice, with a screenshot. No open ticket, no article covers E-409.\n\nTeam lead comment: 'Case 1 has an error code, just escalate it.' (Check 1 matches first.)","rules":[{"type":"plainText"},{"type":"maxWords","max":40}],"mistakes":[{"title":"Escalating the duplicate","why":"The team lead comment points to the error code.","fix":"Say: \"Check in order and stop at the first match.\"","checks":["r1"]},{"title":"Using free-form actions","why":"The AI writes 'Send article and follow up'.","fix":"Say: \"Action is exactly one of the four phrases.\"","checks":["r2"]},{"title":"Extra text","why":"The AI explains each case.","fix":"Say: \"Four lines only: Case N: Action.\"","checks":["r0"]},{"title":"Getting markdown back","why":"AI adds bold and headings by default, which show up as stray symbols where this is pasted.","fix":"Say \"plain text, no bold, no headings, no asterisks\".","checks":["plainText"]},{"title":"Trusting the AI to count words","why":"AI is bad at counting and usually runs long, so a vague \"keep it short\" lands well over the limit.","fix":"Cap each part instead of the total, for example \"each line under 15 words\".","checks":["maxWords"]}],"promptHints":[{"label":"Asks for plain text","patterns":["plain","no bold","no markdown","no formatting","no headings"]},{"label":"Sets a length limit","patterns":["\\d+\\s*words","word (limit|count)","under \\d+"]},{"label":"States the order","patterns":["in order","first match","stop"]},{"label":"Lists the allowed actions","patterns":["resolve now","close as duplicate","ask customer for info"]}],"hint":"Decision trees depend on order and fixed outputs. Say to stop at the first match and list the four exact actions.","rubricLabels":["The output is exactly four lines formatted Case N: Action","The actions are Case 1 Close as duplicate and Case 2 Ask customer for info","Case 3 is Resolve now and Case 4 is Escalate to Tier 2, and every action is exactly one of the four allowed phrases"]}